gd->ram_size is reduced in this function to reserve secure memory.
Avoid running this function again to further reduce memory size.
This fixes issue for SPL boot with PPA image loaded in which case
secure memory is incorrectly allocated due to repeated calling.
Signed-off-by: York Sun <york.sun@nxp.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This change is required due to trusted OS (OP-TEE) not being position
independent code, it requires compile time fixed base address.
To take care of this it is assumed that all layerscape armv8 platforms
has minimum 2G ddr in first region. So we can have fixed address
space (66 MB -> (2MB for PPA + 64MB for trusted OS)) allocated from
top of first 2G ddr region and compile trusted OS with this fixed
base address.
But one exception here is ls1012 where we have only 1G (rdb) or 512M
(frdm) ddr memory. For those we can have different fixed compile time
base addresses for trusted OS.
Signed-off-by: Sumit Garg <sumit.garg@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
In case high region memory doesn't have enough space for Management
Complex (MC), the return value should indicate a failure so the
caller can handle it accordingly.
Signed-off-by: York Sun <york.sun@nxp.com>
Reported-by: Ebony Zhu <ebony.zhu@nxp.com>
This patch adds support for RGMII protocol
NXP's LDPAA2 support RGMII protocol. LS1088A is the
first Soc supporting both RGMII and SGMII.
Signed-off-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
Signed-off-by: Amrita Kumari <amrita.kumari@nxp.com>
Signed-off-by: Ashish Kumar <Ashish.Kumar@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
CoreLink Cache Coherent Interconnect (CCI) provides full cache
coherency between two clusters of multi-core CPUs and I/O coherency
for devices and I/O masters.
This patch add new config option SYS_FSL_HAS_CCI400 and moves
existing register space definaton of CCI-400 bus to fsl_immap to be
shared. CONFIG_SYS_CCI400_ADDR is replaced with SYS_CCI400_OFFSET
in Kconfig.
Signed-off-by: Ashish Kumar <Ashish.Kumar@nxp.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
[YS: revised commit message, squashed patches for armv8 and armv7]
Reviewed-by: York Sun <york.sun@nxp.com>
The declarations should not be in common.h. Move them to the arch-specific
headers.
Signed-off-by: Simon Glass <sjg@chromium.org>
[trini: Fixup thinko defined(FSL_LSCH3) -> defined(CONFIG_FSL_LSCH3)]
Signed-off-by: Tom Rini <trini@konsulko.com>
When U-Boot boots from EL2, skip some lowlevel init code requiring
EL3, including CCI-400/CCN-504, trust zone, GIC, etc. These
initialization tasks are carried out before U-Boot runs. This applies
to the RAM version image used for SPL boot if PPA is loaded first.
Signed-off-by: York Sun <york.sun@nxp.com>
The QorIQ LS2081A SoC has eight 64-bit ARM v8 Cortex A72 cores and
is built on layerscape architecture. It is 40-pin derivative of
LS2084A (non-AIOP personality of LS2088A). So feature-wise it is
same as LS2084A. LS2041A is a 4-core personality of LS2081A.
Signed-off-by: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: Santan Kumar <santan.kumar@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
PSCI can be used on both multiple and single core socs. Current
implementation only allows PSCI to work on multiple core socs.
This patch removes this restriction so that PSCI can work on
single core socs as well.
Signed-off-by: Chenhui Zhao <chenhui.zhao@nxp.com>
Signed-off-by: Tang Yuantian <andy.tang@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
This function name shadows a global name but is in fact different. This
is very confusing. Rename it to help with the following refactoring.
Signed-off-by: Simon Glass <sjg@chromium.org>
By making dram_init_banksize() return an error code we can drop the
wrapper. Adjust this and clean up all implementations.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
At present we cannot use this function as an init sequence call without a
wrapper, since it returns the RAM size. Adjust it to set the RAM size in
global_data instead, and return 0 on success.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
It looks like only cm5200 and tqm8xx use this feature, so we don't really
need it in generic code. Drop it and have the users access gd->board_type
directly.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
The LS2088A series SoCs has different physical memory map address and
CCSR registers address against LS2080A series SoCs.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
Erratum A009635 is valid only for LS2080A SoC and its
personality. Add SoC svr check.
Signed-off-by: Priyanka Jain <priyanka.jain@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
In early MMU table, DDR has to be mapped as device memory to avoid
speculative access. After DDR is initialized, it needs to be updated
to normal memory to allow code execution. To simplify the code,
dram_init() is moved into a common file as a weak function.
Signed-off-by: York Sun <york.sun@nxp.com>
This patch follows the break-before-make process when making changes
to MMU table. MMU is disabled before changing TTBR to avoid any
potential race condition.
Signed-off-by: York Sun <york.sun@nxp.com>
Update mapping with actual DDR size. Non-existing memory should not
be mapped as "normal" memory to avoid speculative access.
Signed-off-by: York Sun <york.sun@nxp.com>
For ARMv8 Layerscape SoCs, secure memory and MC memorey are reserved
at the end of DDR. DDR is spit into two or three banks. This patch
reverts commit aabd7ddb and simplifies the calculation of reserved
memory, and moves the code into common SoC file. Secure memory is
carved out first. DDR bank size is reduced. Reserved memory is then
allocated on the top of available memory. U-Boot still has access
to reserved memory as data transferring is needed. Device tree is
fixed with reduced memory size to hide the reserved memory from OS.
The same region is reserved for efi_loader.
Signed-off-by: York Sun <york.sun@nxp.com>
Moved the config FSL_PPA_ARMV8_PSCI from fsl-layerscape's Kconfig to
Kconfig under armv8 and renamed it to SEC_FIRMWARE_ARMV8_PSCI.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
Up to now, there are 3 kind of SoCs under Layerscape Chassis 2,
like LS1043A, LS1046A and LS1012A. But the clocks tree has a
lot of differences, for instance, the IP modules have different
dividers to derive its clock from Platform PLL. And the core
cluster PLL and platform PLL maybe have different reference
clocks, such as LS1012A. Another problem is which clock/PLL
should be described by sys_info->freq_systembus, it is confused
in Layerscape Chissis 2.
This patch is to bind the sys_info->freq_systembus to the Platform
PLL, and handle the different divider of IP modules separately
between different SoCs, and separate reference clocks of core
cluster PLL and platform PLL.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
NXP/Freescale uses macro CONFIG_ARMV8_PSCI to enable their private PSCI
implementation in PPA firmware, but this macro naming too generic, so this
patch replaces it with a specic one CONFIG_FSL_PPA_ARMV8_PSCI.
And this macro CONFIG_ARMV8_PSCI will be used for a generic PSCI for ARMv8
which will be added in following patchs.
Signed-off-by: Hongbo Zhang <hongbo.zhang@nxp.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: York Sun <york.sun@nxp.com>
- add additional function erratum_a009942_check_cpo to check if the
board needs tuning CPO calibration for optimal setting.
- move ERRATUM_A009942(with revision to check cpo_sample option) from
fsl_ddr_gen4.c to ctrl_regs.c for reuse on all DDR4/DDR3 parts.
- move ERRATUM_A008378 from fsl_ddr_gen4.c to ctrl_regs.c
- remove obsolete ERRATUM_A004934 which is replaced with ERRATUM_A009942.
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@nxp.com>
[YS: Replaced CONFIG_QEMU_E500 with CONFIG_ARCH_QEMU_E500]
Reviewed-by: York Sun <york.sun@nxp.com>
NXP ARMv8 SoC LS2080A release all secondary cores in one-go.
But other new SoCs like LS2088A, LS1088A release secondary
cores one by one.
Update code to release secondary cores based on SoC SVR
Add code to release cores one by one for non LS2080A SoCs
Signed-off-by: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: Raghav Dogra <raghav.dogra@nxp.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
[YS: remove "inline" from declaration of initiator_type]
Reviewed-by: York Sun <york.sun@nxp.com>
Timer controller base address has been changed from
LS2080A SoC (and its personalities) to new SoCs like
LS2088A, LS1088A.
Use SVR based timer base address detection to avoid compile time #ifdef.
Signed-off-by: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
When implementing efi loader support, we can expose runtime services
for payloads. One such service is CPU reset.
This patch implements RTS CPU reset support for layerscape systems.
Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: York Sun <york.sun@nxp.com>
The core position is not continuous for some SoCs. For example,
valid cores may present at position 0, 1, 4, 5, 8, 9, etc. Some
registers (including boot release register) only count existing
cores. Current implementation of cpu_mask() complies with the
continuous numbering. However, command "cpu status" queries the
spin table with actual core position. Add functions to calculate
core position from core number, to correctly calculate offsets.
Tested on LS2080ARDB and LS1043ARDB.
Signed-off-by: York Sun <york.sun@nxp.com>
The exact SoC revsion number can be recognized from U-Boot log.
Signed-off-by: Wenbin Song <wenbin.song@nxp.com>
Signed-off-by: Mingkai Hu <mingkai.hu@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
Now, arch/${ARCH}/include/asm/errno.h and include/linux/errno.h have
the same content. (both just wrap <asm-generic/errno.h>)
Replace all include directives for <asm/errno.h> with <linux/errno.h>.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
[trini: Fixup include/clk.]
Signed-off-by: Tom Rini <trini@konsulko.com>
MMU bit in SCTLR needs to be set explicitly after tables are
created. It isn't an issue for EL3 becuase this bit is already
set by early MMU setup. But for other exception levels this
bit was not set.
Signed-off-by: York Sun <york.sun@nxp.com>
Add support to detect Cortex-A72 core for printing it out.
The Initiator Version of A72 core should be 0x4.
Signed-off-by: Alison Wang <alison.wang@nxp.com>
Signed-off-by: Mingkai Hu <mingkai.hu@nxp.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
If the PSCI and PPA is ready, skip the fixup for spin-table and
waking secondary cores. Otherwise, change SMP method to spin-table,
and the device node of PSCI will be removed.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
This function assume that the d-cache and MMU has been enabled earlier,
so it just created MMU table in main memory. But the assumption is not
always correct, for example, the early setup is done in EL3, while
enable_caches() is called when the PE has turned into another EL.
Define the function mmu_setup() for fsl-layerscape to cover the weak
one.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
Drop platform code to create static MMU tables. Use common framework
to create MMU tables on the run. Tested on LS2080ARDB with secure and
non-secure ram scenarios.
Signed-off-by: York Sun <york.sun@nxp.com>
Secure_ram variable was put in generic global data. But only ARMv8
uses this variable. Move it to ARM specific data structure.
Signed-off-by: York Sun <york.sun@nxp.com>
Adds get_svr and IS_SVR_REV helpers for ARMv8 platforms,
similar to PPC and ARMv7.
Signed-off-by: Sriram Dash <sriram.dash@nxp.com>
Signed-off-by: Rajesh Bhagat <rajesh.bhagat@nxp.com>
The SP805-WDT module on LS2080A requires configuration of PMU's
PCTBENR register to enable watchdog counter decrement and reset
signal generation. The watchdog clock needs to be enabled first.
Signed-off-by: Yunhui Cui <yunhui.cui@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
As the issue about the stack will get corrupted when switching between
the early and final mmu tables is fixed by commit 70e21b064, the
workaround to flush dcache is unnecessary and should be removed.
Signed-off-by: Alison Wang <alison.wang@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
LS2080A is the primary SoC, and LS2085A is a personality with AIOP
and DPAA DDR. The RDB and QDS boards support both personality. By
detecting the SVR at runtime, a single image per board can support
both SoCs. It gives users flexibility to swtich SoC without the need
to reprogram the board.
Signed-off-by: York Sun <york.sun@nxp.com>
CC: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
With commit 7985cdf we converted all systems except for the Layerscape
SoCs to the generic descriptor table based page table setup.
On the Layerscape SoCs however, we just provide an empty table stub
and do the setup ourselves. To reserve enough memory for the tables,
we need to override the default counting mechanism which would end up
with an empty table because we have no maps.
Fixes: 7985cdf
Reported-by: York Sun <york.sun@nxp.com>
CC: Alison Wang <alison.wang@nxp.com>
CC: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Tested-by: York Sun <york.sun@nxp.com>
Reviewed-by: York Sun <york.sun@nxp.com>
By now the code to only have a single page table level with 64k page
size and 42 bit address space is no longer used by any board in tree,
so we can safely remove it.
To clean up code, move the layerscape mmu code to the new defines,
removing redundant field definitions.
Signed-off-by: Alexander Graf <agraf@suse.de>
MC and debug server are not board-specific. Move reserving memory to SoC
file, using the new board_reserve_ram_top function. Reduce debug server
memory by 2MB to make room for secure memory.
In the system with MC and debug server, the top of u-boot memory
is not the end of memory. PRAM is not used for this reservation.
Signed-off-by: York Sun <yorksun@freescale.com>
DDR has been set as secure in MMU tables. Non-secure master such
as SDHC DMA cannot access data correctly. Mixing secure and non-
secure MMU entries requirs the MMU tables themselves in secure
memory. This patch moves MMU tables into a secure DDR area.
Early MMU tables are changed to set DDR as non-secure. A new
table is added into final MMU tables so secure memory can have
2MB granuality.
gd->secure_ram tracks the location of this secure memory. For
ARMv8 SoCs, the RAM base is not zero and RAM is divided into several
banks. gd->secure_ram needs to be maintained before using. This
maintenance is board-specific, depending on the SoC and memory
bank of the secure memory falls into.
Signed-off-by: York Sun <yorksun@freescale.com>
If the core runs at higher than x3 speed of the platform, there is
possiblity about sev instruction to getting missed by other cores.
This is because of SoC Run Control block may not able to sample
the EVENTI(Sev) signals.
Configure Run Control and EPU to periodically send out EVENTI signals to
wake up A57 cores.
Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
